Authors Accepted Manuscript

Novel method for research on exposure to nonlinear

vibration transferred by suspension of vehicle

Rafa Burdzik

www.elsevier.com/locate/nlm

PII: S0020-7462(16)30281-5
DOI: http://dx.doi.org/10.1016/j.ijnonlinmec.2016.10.014
Reference: NLM2726
To appear in: International Journal of Non-Linear Mechanics
Received date: 26 March 2016
Revised date: 23 October 2016
Accepted date: 26 October 2016
Cite this article as: Rafa Burdzik, Novel method for research on exposure to
nonlinear vibration transferred by suspension of vehicle, International Journal of
Non-Linear Mechanics, http://dx.doi.org/10.1016/j.ijnonlinmec.2016.10.014
This is a PDF file of an unedited manuscript that has been accepted for
publication. As a service to our customers we are providing this early version of
the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting galley proof before it is published in its final citable form.
Please note that during the production process errors may be discovered which
could affect the content, and all legal disclaimers that apply to the journal pertain.
 Novel method for research on exposure to nonlinear vibration transferred by suspension of
vehicle
Rafa Burdzik
Silesian University of Technology, Faculty of Transport, Department of Automotive Vehicle
Construction, Krasiskiego 8 Street, 40-019 Katowice, Poland

rafal.burdzik@polsl.pl

Abstract
The paper presents novel method for research on exposure to nonlinear vibration of
passenger car suspension as nonlinear dynamical system. Also paper provides a discussion on
the results of studies addressing the impact exerted by damping properties of shock absorber
on the vibrations being generated. The research was conducted on the car forced to vibration
by the exciter machine with changeable frequency. The paper addresses results of analysis of
application of one of the time-frequency representation techniques to the identification of
structure of vibration. The obtained representation of the vibration allows determining time
function of separate frequency bands, which represents the isolated vibration dynamics
phenomena. Considering the variability of the time and frequency distribution of the vibration
even in selected analyzed bands, the time function was developed as exposure to vibration
estimator. The recommended method makes use of function of exposure to vibration. The
advantage of these method is possibilities of precise analysis of chosen frequency bands. The
investigations conducted confirm considerable susceptibility to changes in the technical
condition of shock absorber with regard to the assessment of human exposure to vibrations.

Graphical Highlights

Vibration of the excitation platform

Scheme of the experimental research

 arm

car-body

Comparison of spectrums of unsprung and sprung masses

Time-frequency distribution of the vibration of suspension arm (shock absorber with: left - 100%, right - 50%
of liquid volume)
fb N 1 2
j(
x(t ) g (t s) e
k )t
St f N

f fa t 0

Time-frequency distribution of the vibration of suspension arm for chosen frequency bands (shock absorber
with: left - 100%, right - 50% of liquid volume)
fb
1
Savr ( S ( f , ti ))
( fb f a ) / d
tf
i 0 f fa
a) b)
The process of the determination of the average value of frequency bands vibration (11 13 Hz) in time domain:
a) suspension arm, b) upper mounting of shock absorber

Time function of average vibration in frequency bands: left- 20.5-22.5 Hz, right- 63.5-65.5

Keywords: exposure to vibration, vehicle vibration, damping characteristics.

1. Introduction
The paper refers to presentation on 13th International Conference on Dynamical Systems
Theory and Applications. Author presented application of own original method of vibration
analysis for investigation on passenger car suspension as nonlinear dynamical system [1].
This paper is extension of this method due to issue of exposure to non-linear vibration on
passenger car example.
Vibration is the mechanical phenomena caused by machines in operation. Generally the
vibration is undesirable, wasting energy and creating unwanted effects. The vehicle vibrations
are one of the most important unwanted effects. Vibration also causes decrease of safety and
comfort factors and increase of fuel consumption. The range of impacts vibrations exert on a
vehicle driver is very broad, starting from the feeling of discomfort to safety hazards caused
by vibrations at resonant frequencies of specific organs, thus affecting the drivers responses.
Ride quality and comfort increasingly begins refer to vibration and noise in terms of sensation
or feel of the passengers. Lower range of vibration frequencies mostly are correlated to
vibration comfort and higher (more than 80 Hz) to noise [2]. Vibration and noise are
perceived differently by humans, so there is a need to adopt methods that help quantify and
control them [3,4]. Independent of method it is important to define the sources of these
phenomena. The analysis of the way in which the structure responds to the forcing, and in
particular the way the vibration propagates from a source into distant parts of the system
allow to research on vibration phenomena in means of transport. There is more than one
approach to the theory of structural vibration, and it is useful at the outset to recognise a basic
division into three general classes. These can be characterised as the waves, modes and
rays/wavepackets approaches. All three have advantages for particular types of problem
[5,6].
The identification of sources and propagations path of dynamics phenomena requires
analysis in terms and conditions which are determining the course, modes and perception of
the phenomena. The vibration has to be considered in terms of amplitude, time and frequency.
The evaluation of impact of chosen factors on vibration occurring in motor vehicles requires
wide scope of amplitude, time and frequency characteristics analysis. The properties of
human vibration perception require observation of distribution of vibration in time and
frequency domains for the analysis of vibration phenomena in mean of transport. Thus it is
very helpful to identify structure of the vibration as correlated time-frequency distribution.
Such an approach enables the analysis of exposure to vibration. This presentation of
vibrations structure enables identification of time of exposure to chosen vibration dynamics.
This approach allows precise evaluation of exposure to vibration correlated to natural
vibration of human organs. In the scope of engineering applications the presented approach
allows designing and developing of elements and systems dedicated to absorb or isolate of
vibration in chosen frequency bands.
A very large group of people exposed to general vibrations comprises car drivers,
passengers, tram drivers or building and road machinery operators. Occupations that require
driving long distances or operating heavy equipment expose workers daily to low-frequency
vibrations generally less than 100 Hz. Exposure to these vibrations can cause serious physical
problems ranging from chronic back pain to nerve damage. Whole-body vibration is caused
by twisted sitting postures combined with vibration. The combination increases stress and
load on the neck, shoulder and lower back.
One may think of numerous systems and components of a vehicle which purpose is to
absorb vibrations or minimise their impact. A vehicle suspension system, which consists of
damping, springing and steering elements, are responsible to a considerable extent for
damping of vibrations generated by road irregularities [7-11]. Rubber components which
provide mounting to such systems as engine, gearbox or exhaust system also bear a fair share
of responsibility for absorption of the vibrations generated by these assemblies. The purpose
of every shock absorbing system and element is to minimise the impact exerted by vibrations
on other systems as well as on the persons inside the vehicle. Oscillatory waves propagate and
influence people via the vehicle structure, i.e. the frame or the body. General vibrations are
mainly transferred on men via lower extremities as well as the middle and lower section of the
spine. The growing use of vehicles in urban areas causes that comfort in low-speed driving
phase is an increasingly important consideration for manufacturers. The main sources of
constraints in the occupants environment are vibrations and sounds transmitted to the cabin
via floor pan, dashboard, seats and steering wheels [12,13]. If we consider the wave
propagation in solid structure of car body it is important to analyse the material properties and
jointing techniques [14-17,21].
Thus the paper provides a discussion on application of novel method for research on
exposure to non-linear vibration transferred by suspension of vehicle.

2. State of art

2.1 Vehicle vibration

Vehicle vibrations are classified into two general types of the rigid body vibration of
sprung and unsprung mass and stiffness vibration. Stiffness vibration is determined by mass,
stiffness and damping forces which are contained in suspension component parts.
When the frequency of force usually acting on a vehicle matches or exceeds its natural
oscillations, a vibration resonance is caused and the amplitude is increased, resulting frequent
failures of the suspension and other elements of vehicle. During the drive it can cause wheels
lose grip with the ground in effect vehicle becomes uncontrolled. Thus the vibration transfer
from the road to the occupants can be illustrated as transmittance by multi-filter element. The
excitation is given by the wheel movement caused by the road roughness. First filtering of
vibration occurs as result of stiffness and damping properties of tire. Second and third will be
results of stiffness and damping properties of suspension and car-body. The last filter is
biodynamics properties of human body. This path of propagation determined the human
exposure to vibration and feeling of discomfort. The characteristic of the spring and shock-
absorber in the vehicle suspension system as well as other vehicle elastic elements are
nonlinear. Therefore, vehicle nonlinear oscillatory models are used in oscillatory behavior
analyses [18].The paper [19] presents results of the investigation on influence of air spring
and hydraulic shock absorber of the bus drivers seat on driver comfort.
Vibrations are also a very ample source of information on the technical condition, and
hence they are commonly used in diagnostic systems [20-27]. Bearing in mind such extensive
fields of influence and application of vibrations, it is highly recommendable that the impact
exerted by selected parameters on the form of vibrations recorded as well as their distribution
in the vehicle structure should be studied. This assumption extend potential application of
results in vehicle vibration research, for many purposes, e.g. vibration exposure, vibration
based diagnostics systems and so on. Thus the developed method has been tested on different
damping properties of shock absorber.
The results obtained while studying the impact of the technical condition of the suspension
system components on vibrations, and special emphasis has been put on describing how
passengers are affects through feet. The purpose of the said studies was to identify factors
affecting the vibrations caused by the dynamic impact of road irregularities on wheels of a
moving vehicle. The aspects analysed included the accelerations of vibrations of the
suspension system components generated by the kinematic input function and affecting the
vehicle wheel.

2.2 Vibration testing

Vibration analysis and dynamic testing traditionally rely either on time-domain or

frequency-domain approaches. However, many engineering systems exhibit time-variant
behavior [28]. To meet warranty and reliability requirements demanded by consumers in the
automotive industry many of application of uniaxial and multiaxial excitation for durability
and qualification testing have been developed. In case of vibration safety and comfort a
typical automobile prototype is subjected to a variety of vibration profiles that simulate
operational conditions during R&D stage [29]. Vibration tests are also used to utilize
accelerated life testing for critical components to meet the life-cycle requirement while
reducing the testing cost [30]. Another group of approach of vibration testing for validating
vehicle durability relies on is the four-post hydraulic shaker. It is a relatively simple and cost
effective configuration (for automotive manufacturers budget) for performing a complete
vehicle vibration evaluation. It should be noted that using a four- or twelve-hydraulic actuator
shaker introduces some limitations [31]. The first limitation is the inability to simulate the
effect of rolling tires on the overall stiffness and damping of the suspension system and
vehicle structure. The second issue is that hydraulic shakers have limited frequency range
[30], in general 180Hz due to ISO requirements for analysis of human exposure to whole-
body vibration [32].
Researchers are still struggling to system with linear and nonlinear structures under
multiaxial loading. E.g. vehicle life-cycle prediction under multiaxial dynamic loading has
been shown to be extremely complex and more intractable than uniaxial models [33].
Taking into consideration all mechanisms and processes due to complex vibration loading
is becoming very important for current systems that are progressively complex. E.g. failure
mechanisms can be exploited through accelerated multiaxial vibration testing, which may
provide reliability improvements [34].
2.3 Application of time-frequency representation methods for determination of vibration
structure
Signal analysis addresses the problem of extracting information from registered signal and
converting it into a recognizable form. The time courses recorded provide information on the
phenomenon in question, however, in order to extract it, application of the appropriate
mathematical method proving is necessary. The vehicle vibrations belong to a group of non-
stationary signals the analysis of which enforces one to apply time-frequency methods. For
the non-stationary signal, when the components of the signal changes with time, the
representation of the signal in frequency domain become less useful. Simultaneous extraction
of information concerning the time-frequency structure of a signal being analysed is possible
owing to 2- dimensional transforms. To achieve frequency representation that includes
dependence on time the double arguments function is needed, i.e. time-frequency
representation (TFR).
The first attempt to develop such a function was carried out by Gabor in 1946 [35]. His
approach can be considered as special case of the Short-Time Fourier Transform (STFT),
expressed as:

S(,b) =

x(t) w ,s (t) dt =

x(t) g(t s) e -j t dt (1)

where:
s window displacement,
analysing frequencies,
j imaginary unit,
g(t - b)=const constant width of successively analysed window.
The g(t - s) window is typical chose as the Gaussian function or Hann function. The good
properties of this window function occur from smoothly and symmetrically falls to zero
around the time t=0 and is centered about t=s. The STFT segmenting the signal into narrow
time intervals and takes the Fourier transform of each segment. Thus each Fourier transform
provides the spectral information of a separate time-slice of the signal, providing
simultaneous time and frequency information. Some disadvantage of STFT is a constant
width of the window. If the window is too long transformation providing excellent frequency
localization, but no time information. When the window is too short transformation gives the
time signal back, with a phase factor, providing excellent time localization but no frequency
information.
Simultaneous extraction of information concerning the time-frequency structure of a signal
being analysed is possible owing to a wavelet transform. It is so because the wavelet function
applied in this transformation can be extended or shortened. Narrow wavelets make it possible
to analyse high-frequency components of a signal, and appropriately long base functions
reveal its slowly-variable properties. The template role is performed by basic wavelet (t).
The wavelet functions as a transformation kernel. A single wavelet is used in the given
transformation, however, due to modification of scale coefficient a and modification
coefficient b, it forms what is referred to as a wavelet family. A wavelet transformation of
signal x(t) may be defined as follows:

1 t b
WTx (a, b)
a
x(t )
a
dt (2)

where:
(t) wavelet family,
(t b) / a transformation kernel,
1
a scaling parameter effecting frequency change a R / 0 a ~ , f frequency,
f
1
- wavelet standardized constant,
a
b shift parameter locating the wavelet position against time axis b R .
 The wavelet transform represents the correlation between the signal analysed and function
(t) having been appropriately scaled. The very idea of wavelet transformation is the
decomposition of signal x(t) into wavelet coefficients WTx(a,b) by application of the base
function. As a result of such a transformation, one obtains coefficients being functions of
scale and time. By altering the parameters of scale a and time shift b, one may obtain the
time-frequency distribution. The base function of the wavelet transformation is subject to the
operations of scaling and shifting which makes it possible to obtain wavelets of various
lengths of decay time and centre frequencies. The best results are obtained by this method
when wavelets adjusted in shape to the relevant signal features are used.
The Wigner-Ville distribution approach is based on the use of the autocorrelation function
for calculating the power spectrum. To construct the autocorrelation function, the signal is
compared to itself for all possible relative shifts, or lags. The Wigner-Ville uses a variation of
the autocorrelation function where time remains in the result, called instantaneous
autocorrelation function. The Wigner-Ville Distribution (WVD) is defined as:
j 2 t

WVDxx ( , ) x * (t 2 ) x(t 2 ) e dt (3)
where:
x* represents the complex conjugate of the signal x,
time shift,
j imaginary unit,
frequency shift.

The analytic associate x(t) of a signal s(t) is defined such that x(t) s(t)+iH[s(t)], where
H[s(t)] is the Hilbert Transform of the signal s(t). In most cases, the TFR techniques are
applied to the analytic associates of real signals unless noted otherwise in particular, x(t) is
generally the complex-valued analytic associate of some real-valued time signal of interest.
WVD is obtained by comparing the information of the signal with its own information at
other times and frequencies. The WVD is a quadratic TFR, and most other TFR methods
(including the spectrogram, i.e. STFT and the scalogram, i.e. WT) can be derived from the
WVD, with a suitable choice of smoothing factors.
Wigner distribution was first introduced by E. Wigner in the context of quantum mechanics
(Wigner, 1932), and later independently developed by J. Ville who applied the same
transformation to signal processing and spectral analysis (Ville, 1948) [36,37].

2.4. Novel approach on vibration research due to exposure

The current state of art presents many different approaches and methods for vehicle
vibration signal processing. Studies presented in [38] focused on the dynamics of a passive
car suspension, the nonlinear characteristics of tyres and the effect of shimming in vehicle
wheels. However, due to various nonlinearities in the vehicle dynamics, the chaotic behaviour
may produce noise like responses [39]. Authors of [40] have studied the vibrations of a
quarter-car model with a softening stiffness of the Duffing type, focusing on the potential for
chaotic behaviour. They examined the global homoclinic bifurcations that appear as
instabilities at the boundaries of the basins of attraction, and the cross-sections of stable and
unstable manifolds, by the perturbation approach using Melnikov theory. A critical amplitude
was found for which the system can exhibit chaotic vibrations. The paper [41] presents a
nonlinear decoupling approach based on the Modified Generalized Frequency Response
Functions and the nonlinear feature of phase in variance, for the pure nonlinearity input
nonlinear system. When nonlinearity is introduced into linear system, nonlinear behaviour is
the rule in the dynamic behaviour of the physical system [42].
Vehicle vibration has to be considered in terms of safety and comfort. Shock absorber is
one of the most important element of suspension. Thus periodic test of shock absorber
damping are obligatory in many countries. When consider vehicle as multibody mechanical
system, safety condition requires constant grip of wheel and road roughness. Hence vibration
occurring for unsprung masses resonance band are determinant of vibration safety factor. For
the analysis of vibration comfort it is important to determine the exposure time. The most
popular standards for measurement and evaluation of human response to Whole-Body
vibration (WBV) are ISO 2631-1 (1997), ISO2631-5 (2004) and British Standard 6841
(1987). The basic evaluation method makes use of the RMS acceleration aw. It is unclear if
worst axis acceleration or triaxial vector sum has to be used for evaluating the vibration. Also
six frequency curves are presented covering the most significant combinations of vibration
axes and effects (health, comfort, perception, motion sickness). Three of these curves are used
in the evaluation of vibration from the point of view of health effects, but only two in the
procedure of assessment of vibration severity. Another method use of the time integrated
fourth power of acceleration, known as Vibration Dose Value (VDV). This value is however
unjustified. It is not specified how the total triaxial VDV can be calculated from the axial
VDVs. When only RMS measurements of aw are available [43]. Both standards assume that
acceleration magnitude, frequency spectrum and duration represent the principal exposure
variables which account for the potential harmful effects. One of the alternative method is the
possibility of using the power absorbed (PABS) by the human body when exposed to seat
transmitted vibration, in order to evaluate and assess the vibration, was first explored by a
group of US Army engineers in the mid-60s. The power absorbed by the subject exposed to
vibration is calculated from the simultaneous measurement of force, acceleration and their
relative phase [43,44]. WBV is field of investigation of many scientists of which the most
significant achievements have M.J. Griffin [49-51].
Vibration assessment depends strongly on the duration of exposure. Thus author proposed
novel approach for analysis of exposure to vibration. This method can be adopted for the
research on vibration safety or comfort, especially for the vehicle vibration. Using of time-
frequency representation (TFR) of the signal allows precise analysis and evaluation duration
of exposure for defined frequency bands. The proper selection of the frequency enables the
analysis of safety. For such purpose the resonance frequency band of unsprung masses (ca.
10-16 Hz) has to be observed. For the analysis of vibration comfort it is required to select
frequency of natural vibration of chosen human organs. This method enables the
determination of exposure function. The calculation method and exposure function have been
tested by the experimental investigation on the real car. Additionally the suspension of the car
was equipped with shock absorber with strongly different damping characteristics, as
significant element of car considered as vibration model.

3 Research methodology

One of the main sources of vehicle vibration is vertical movement of the wheel. It is
caused by driving on road roughness. The excitation can be considered as time functions of
the wheel strokes. It causes force transfer and dynamics response to the vehicle construction.
The analysis of the mechanical system dynamics requires conducting research with vibration
excitation system. It enables forcing of the tested system with the excitation programmed for
the chosen frequency bands. One of the assumptions of the research method was guarantee of
the force excitation recurrence with possibilities of changing of the technical parameters of
the researched vehicle.
 The goal of the investigation was to develop and verify novel method for research on
exposure to nonlinear vibration transferred by suspension of vehicle.
The research comprised a series of active diagnostic experiments conducted in passenger
cars. A car prepared to the tests, with the suspension components mounted having their
technical condition known was subject to oscillatory input functions. For that purpose, a
station enabling kinematic vibration excitation was used, making it possible to programme the
dynamic frequency structure of input functions. The input frequency range envisaged
comprised resonance frequency bands of both sprung and unsprung masses. The distribution
of the vibration of excitation platform affecting on the wheel has been depicted in Figure 1.

Fig. 1. Vibration of the excitation platform

Each time the test was conducted, a wheel of one vehicle axle was subject to the input
function. For the sake of analysis of the suspension system damping parameters, vertical
vibrations of excitation platform, unsprung elements (suspension arm) and the vehicle body
(upper shock absorber mounting) were recorded coaxially (fig. 2). The scheme of the
experimental research has been depicted in Figure 3.

Fig. 2. Arrangement of vibration sensors across the suspension and car body of the vehicle
tested
Fig. 3. Scheme of the experimental research

One of the most important components of the vehicle suspension is shock absorber. It is
responsible for damping of the vibration transferred from wheel into car body [20]. Thus for
the purpose of verification of the method the investigations were conducted for different
damping properties of the shock absorber. The damping properties of the shock absorber are
described as force vs. velocity characteristics. One of the most common reason of decrease of
damping properties during operating of vehicle is leak of shock absorber fluid. The influence
of the volume of the shock absorber fluid on damping characteristic has been depicted in
Figure 4. It is very significant how the damping force changed under different fluid volume.
When consider novel suspensions with active vibration control such kind of characteristics are
not enough to represent damping properties [45].
The paper presents results occurring during research vehicle with shock absorber with
damping characteristic presented as force vs. velocity functions depicted in Figure 4.

Fig. 4. Force vs. velocity characteristics of the investigated shock absorbers

As the confirmation of significant impact of shock absorber on damping of vibration

transferred from unsprung to sprung masses of the vehicle the comparison of FFT (Fast
Fourier Transformation) of vibration registered on suspension arm and upper mounting of
shock absorber on car body has been depicted in Figure 5 [52]. It can be observed the
minimization of vibration, especially for the resonance frequency bands.

arm

car-body

Fig. 5. Comparison of spectrums of unsprung and sprung masses (evaluation of the damping
properties)

The relationship between the sprung and unsprung masses is very important issue during
design of suspension system. It determines dynamical properties and stability of vehicle. The
proper relationship is always result of compromise between safety and comfort, thus it is
different for sport car, limousine and off road vehicle or any other. Thus it is difficult to define
the optimal ratio between the sprung and the unsprung masses.

4 Research results

Vehicles vibrations are random processes as results of vide range of possibilities forces and
nonlinear characteristics of suspensions elements of the vehicles construction. The vibration
occurring during research are results of forces acting on wheel by the exciter machine. The
signals have non-stationary properties (fig. 6). Thus the evaluation of human vibration
exposure based on global estimators is difficult. Distribution of the vibration in defined
frequency bands enables observing of vibration exposition in time function.
The effects of nonlinearity can be explained as follows: as the resonance causes the
amplitude of the motion to increase, the relation between period and amplitude (which is a
characteristic effect of nonlinearity) causes the resonance to detune, decreasing its tendency to
produce large motions. The change in stability can be observed by using Lindstedts method
in order to obtain an approximation for the amplitude and frequency of the resulting periodic
motion. Some dynamical processes are modeled as differential-delay equations. In the case
that the delay takes on its critical value, it is necessary to take into account the effect of
nonlinear terms [53]. Another method the Routh-Hurwitz stability criterion is a test on the
coefficients of a characteristic polynomial to determine whether the system is linearly stable
[54].
The research object as technical system is example of nonlinear continuous or multiple-
degree of freedom system subject to a general excitation, its response consists of multiple
coupled basis functions. Nonlinearities can be due to vehicle suspension system properties as
large elastic deformations, material properties (i.e., material nonlinearity), backlash between
parts, clearances between mounting brackets, geometric constraints on deformation, dry
friction, and nonlinear hysteretic damping (e.g., damping of shock absorbers fig. 4) [55].
Depending on the characteristics of the vibration signal or even its periods different
mathematical transformations can be made. It enables elaborating of dedicated signal
processing method for research on vibration in many terms. The evaluation of impact of
chosen factors on vibration occurring in motor vehicles requires wide scope of amplitude,
time and frequency characteristics analysis.
The analysis of spectrum is very difficult because frequency components are blurry (fig. 6).
For the constant frequency the signal is stationary and frequency component are strongly
isolated. As the applications of frequency analysis in automotive industry are typically
automotive fatigue research relied on power spectral densities (PSDs) to represent stochastic
dynamic behaviors [46,47]. Due to limitation in evaluation based on the one-dimension
(frequency) analysis, it has been proved that tests and analytical tools are not always
successful in producing failures observed in the field. The Ford Corporation found that a
module could pass the qualification test but fail in the field [48]. Thus the research was
conducted on develop new approach for multidimensional analysis of vibration in vehicle
research.

Fig. 6. Non-stationary signal of vibration of upper mounting of shock absorber on car body

The time-frequency representations of the structure of the vibration enable selective

evaluation of exposure to vibration in means of transport. For the purpose of analysis of the
influence of the technical condition of suspension elements on the vibration propagation it is
useful to identify the components of the signal which are best representation of those
influences. Thus the symptoms of the technical conditions or value of the operating
parameters can be determined.
For the results analysis the STFT of the signals were calculated. The reason of choosing
the STFT was time consumption of the analysis process, because utility and applicability
requirements. The distribution in time of vibration energy for next frequencies can be
observed. For the purpose of better analysis all results were presented as distribution of
absolute values of STFT. The STFT transformation of vibration registered during research on
vehicle with build in shock absorbers with 100% and 50% of liquid volume are shown in
figures below. It shows that for the shock absorber with 50% of liquid the vibration energy
become more centered and occurred in non-stationary parts of the signal.
Fig. 7. Time-frequency distribution of the vibration of suspension arm (shock absorber with:
left - 100%, right - 50% of liquid volume)

Fig. 8. Time-frequency distribution of the vibration of upper mounting of shock absorber

(shock absorber with: left - 100%, right - 50% of liquid volume)
5. Discussion
5.1 Distribution of the vibration of defined frequency bands
The properties of human vibration perception require observation of distribution of
vibration in time and frequency domains for the analysis of vibration phenomena in mean of
transport. Thus it is very helpful to identify structure of the vibration as correlated time-
frequency distribution. Such an approach enables the analysis of exposure to vibration. This
presentation of vibrations structure enables identification of time of exposure to chosen
vibration dynamics. This approach allows precise evaluation of exposure to vibration
correlated to natural vibration of human organs. The paper presents investigations conducted
to develop novel method for research on exposure to vibration transferred into the passenger
car occupants.
The paper addresses results of analysis of application of one of the TFR techniques to the
identification of structure of vibration to identify impact of damping properties on exposure to
vibration. The possibilities of identification of characteristics components of vibration
structure and sensitivity on the changes in mechanical system parameters have been
investigated. The influence of the damping of shock absorber can be observed in vibrations
structure, as an increase of the vibration energy. The time-frequency distribution of the
vibration allows localization of window with highest increase of energy generated during
passing by resonance frequency of unsprung masses in the time period of coasting of exciter
machine.
 Basing on the analysis of the spectrum and TFR results the main frequency bands were
chosen for further observation [52]. It allows to determine frequency values fa and fb, that set
analysed frequency bands. The obtained representation of the vibration as structure of STFT
transformation allow to determine time function of separate frequency bands which represents
the isolated vibration dynamics phenomena, expressed as:
fb N 1 2
j(
x(t ) g (t s) e
k )t
St f N
(4)
f fa t 0

where:
f frequency band from fa to fb,
t time,
N window length,
k = 0,1,2,,N-1 index along the frequency axis.

This function allows separation of the chosen dynamics structure of the vibration and
analysis of energy changes by the time realization in isolated frequency bands.
This method of identification of vibration dynamics allows observing time of increase of
vibration activity of the vehicle and influence of the damping properties on time function of
the chosen frequency bands of the signal. Results analysis of vibration structure, presented in
previous section, allow to assumed that the most sensitivity frequency components are ca. 12
Hz, 21.5 Hz and 64.5 Hz. Thus these frequencies were set as centers of the analysed
frequency bands range ca. 2 Hz each. Figures below present time distribution of the vibration
in these frequencies.

Fig 9. Time-frequency distribution of the vibration of suspension arm for chosen frequency
bands (shock absorber with: left - 100%, right - 50% of liquid volume)

Fig 10. Time distribution of the vibration of upper mounting of shock absorber for chosen
frequency bands (shock absorber with: left - 100%, right - 50% of liquid volume)
5.2 Function of exposure time to the vibration of defined frequency bands
The procedure as regulated by the international ISO 2631-1 standard [32] prescribes the
vibration total value of weighted root mean square accelerations as a reference quantity for
the assessment of vibration effects on the comfort. Many more estimators for the assessment
of vibration effects on the comfort have been presented in [48-51], especially for the child
seat.
Considering the variability of the time and frequency distribution of the vibration even in
selected analyzed bands, the time function was developed as vibration exposure estimator. It
is calculated of average value of frequency bands vibration in time domain, expressed as:
fb
1
Savr ( S ( f , ti )) (5)
i 0 ( fb f a ) / d
tf
f fa
where:
fa lover edge of the passband,
fb upper edge of the passband,
d STFT frequency resolution (integer rounded value),
(fb-fa)/d returns number of samples in frequency bands.

Figures below present process of the determination of the average value of frequency
bands vibration 11 13 Hz, in time domain.
The figures below show the comparison of the time function of average vibration in those
frequency bands for the car with shock absorber with 100% and 50% of liquid volume.

a) b)

Fig. 11. The process of the determination of the average value of frequency bands vibration
(11 13 Hz) in time domain: a) suspension arm, b) upper mounting of shock absorber

Figures below present the comparison of time function of average vibration in chosen
frequency bands for the vehicle with build in shock absorber with 100% and 50% of liquid
volume. The paper presents result obtained for the following frequencies bands: 11-13 Hz
(fig. 11), 20.5-22.5 Hz and 63.5-65.5 Hz (fig. 13,14).
Fig 12. Time function of average vibration in frequency bands: left- 20.5-22.5 Hz, right- 63.5-
65.5 Hz: suspension arm (shock absorber with: blue- 100%, green- 50% of liquid volume)

Fig 13. Time function of average vibration in frequency bands: left- 20.5-22.5 Hz, right- 63.5-
65.5 Hz: upper mounting of shock absorber (shock absorber with: blue- 100%, green- 50%
of liquid volume)

Previous section presents the application of STFT 3-D time-frequency distributions for
proper observation of the vibration structure. The obtained representation of the vibration
allows determining time function of separate frequency bands represents the isolated vibration
dynamics phenomena. The changes of the vibration energy in chosen frequency bands can be
observed by the time function of average vibration in chosen frequency bands. The results of
comparison of the vibration of the vehicle with build in shock absorber with 100% and 50%
of liquid volume show the influence of damping properties on chosen frequency bands.

6 Conclusions

Exposure to vibration in means of transport depends on many factors. For the passengers it
is important what type of vibration penetrates to the human body via vehicle construction or
elements. For the perception properties as type of vibration we can consider the value,
dynamics and exposure time of chosen frequency components of the vibration. The
possibilities of simultaneous observation of distribution of signal in time and frequency
domains are reached by the time-frequency representation (TFR) of the signal.
Such a wide range of experimental studies and analyses enables accurate assessment of the
human exposure to vibrations in vehicles based on multidimensional analysis of the vibration
exposure time for successive frequency bands. It may also initiate a new approach to
experimental modelling and identification of the oscillatory wave propagation in a vehicle
structure.
 The recommended method makes use of function of exposure to vibration. The advantage
of these method is possibilities of precise analysis of chosen frequency bands. The
investigations conducted confirm considerable susceptibility to changes in the technical
condition of shock absorber with regard to the assessment of human exposure to vibrations.
The analysis of the results show increase of the vibration energy for the vehicle with shock
absorber with 50% of liquid volume but for determined time realization. For the rest of the
time of excitation the vibration are even lower, especially during constant excitation (ca.
between 9-14 seconds). The maximums values of the functions are in ca. 5th and 25th seconds
of the excitation. It is the effect of time of passing thru unsprung masses resonance, ca. 12 Hz.
Very important conclusion is that function proposed of exposure time to the vibration of
defined frequency bands is sensitivity for vibration of the suspension element and car body.
The car body vibrations are transfer from vehicle via feet to the humans in mean of transport.
Thus formulated and developed function may be useful for exposure to vibration evaluation or
monitoring.
In the scope of engineering applications the presented approach allows designing and
developing of elements and systems dedicated to absorb or isolate of vibration in chosen
frequency bands.

References

[1] R. Burdzik, Application of original method of vibration analysis for investigation on

passenger car suspension as nonlinear dynamical system. In Awrejcewicz, J. et al. (Eds.):
Dynamical system Mathematical and Numerical Approaches, pp. 127-138. d, (2015)
[2] M. Paweczyk, Acoustics and vibration as fascinating challenges to specialists in
automatic control and signal processing, International Journal of Acoustics and
Vibrations, Vol. 11, Issue 3, 2006, p 118-119.
[3] T. Wszoek, M. Kaczyski, Automatic detection of long-term audible noise indices from
corona phenomena on UHV AC power lines, Acta Physica Polonica A, vol. 125, Issue
4A, pp. A93-A98, 2014.
[4] M. Kaczyski, T. Wszoek, Artificial intelligence and learning systems methods in
supporting long-term acoustic climate monitoring, Acta Physica Polonica A, vol.
123, Issue 6, pp. 1024-1028, 2013.
[5] Z. Dbrowski, M. Zawisza, Investigations of the Vibroacoustic Signals Sensitivity to
Mechanical Defects not Recognised by the OBD System in Diesel Engines. Solid State
Phenomena: Mechatronic Systems, Mechanics and Materials, 180, 2012, p. 194-199.
[6] C.H. Hodges, J. Woodhouse, Theories of noise and vibration transmission in complex
structures, Rep. Prog. Phys., 49, 1986, p. 107-170.
[7] M. Borowiec, J. Hunicz, A. Sen, G. Litak, G. Koszalka, A. Niewczas, Vibration of a
vehicle excited by real road profiles, Forschung im Ingenieurwessen, 74, 2010, p. 99
109.
[8] R. Burdzik, Research on structure and directional distribution of vibration generated by
engine in the location where vibrations penetrate the human organism, Diagnostyka,
14(2), 2013, p. 57-61.
[9] R. Burdzik, Identification of structure and directional distribution of vibration transferred
to car-body from road roughness, Journal of Vibroengineering Vol. 16, Issue 1, 2014, p.
324-333.
[10] R. Burdzik, . Konieczny, Application of vibroacoustic methods for monitoring and
control of comfort and safety of passenger cars, Solid State Phenomena, 210, 2014, p. 20-
25.
[11] S. Wierzbicki, Laboratory control and measurement system of a dual-fuel compression
ignition combustion engine operating in a cogeneration system, Solid State Phenomena,
Vol. 210, pp.200205, 2014.
[12] M. Amari, E. Parizet, V. Roussarie, Multimodal approach to automobile driving comfort:
The influence of visual setting on assessments of vibro-acoustic comfort in simulators,
Applied Acoustic, 74, 2013, p. 13781387.
[13] L. Cho-Chung, C. Chi-Feng, A study on biodynamic models of seated human subjects
exposed to vertical vibration, International Journal of Industrial Ergonomics, 36, 2006, p.
869890.
[14] T. Wgrzyn, The influence of nickel and nitrogen on impact toughness properties of low
alloy basic electrode steel deposits, International Offshore and Polar Engineering
Conference Proceedings, pp. 282-285, 2001.
[15] G. Golanski, J. Jasak, J. Slania, Microstructure, properties and welding of T24 steel -
critical review, Kovove Materialy-Metallic Materials, 52(2), 2014, p. 99-106.
[16] A. Lisiecki, Welding of titanium alloy by Disk laser. Proceedings of SPIE, Laser
Technology, Applications of Lasers, 87030, 2013.
[17] G. Siwiec, B .Oleksiak, A. Smalcerz, J. Wieczorek, Surfacetension of Cu-Ag alloys,
Archives of Materials and Metallurgy, vol. 58(1), pp. 193-195, 2013.
[18] L. Dai, Q. Han, Stability and Hopf bifurcation of a nonlinear model for a four-wheel-
steering vehicle system, Communications in Nonlinear Science and Numerical
Simulation, 9.3, pp. 331-341, 2004.
[19] D. Sekulic, V. Dedovic, S. Rusov, S. alinic, A. Obradovic, Analysis of vibration effects
on the comfort of intercity bus users by oscillatory model with ten degrees of freedom,
Applied Mathematical Modelling, 37, 2013, p. 86298644.
[20] R. Burdzik, . Konieczny, Research on structure, propagation and exposure to general
vibration in passenger car for different damping parameters, Journal of Vibroengineering,
15(4), 2013, p. 1692-1700.
[21] R. Burdzik, Z. Stanik, J. Warczek, Method of assessing the impact of material properties
on the propagation of vibrations excited with a single force impulse, Archives of
Materials and Metallurgy, 57(2), 2012, p. 409-416.
[22] P. Mroz, P. Gorny, M. Ekiert, T. Uhl, An estimation of lithiumion battery state of health -
ageing-included modelling and experimental studies, Diagnostyka, vol. 17, issue 2, 2016,
p. 2534.
[23] D. Dbrowski, W. Batko, W. Cioch, Model of the gears based on multibody system and
its validation by application of non-contact methods, Acta Physica Polonica A, vol. 123,
Issue 6, pp. 1016-1019, 2013.
[24] J. Pankiewicz, P. Deuszkiewicz, J. Dziurd, M. Zawisza, Modeling of powertrain system
dynamic behavior with torsional vibration damper, Advanced Materials Research, vol.
1036, pp 586-591, 2014.
[25] A Wyomaska, J. Obuchowski, R. Zimroz, Procedures for Decision Thresholds Finding
in Maintenance Management of Belt Conveyor System Statistical Modeling of
Diagnostic Data, Proceedings of the 12th International Symposium Continuous Surface
Mining - Aachen 2014, Lecture Notes in Production Engineering, pp 391-402, 2015.
[26] A. Puchalski, I. Komorska, Application of vibration signal Kalman filtering to fault
diagnostics of engine exhaust valve, Journal of Vibroegineering, Vol. 15, Issue 1, pp.152-
158, 2013.
[27] R. Dolecek, J. Novk, O. Cern, M. Bartomiejczyk, Interference in power system for
traction drive with PMSM, Przegld Elektrotechniczny, R. 88(4), pp. 204-207, 2012.
[28] K. Dziedziech,W. J. Staszewski, and T. Uhl, Wavelet-based modal analysis for time-
variant systems, Mechanical Systems and Signal Processing, vol. 50-51, pp. 323-337,
2015.
[29] Ed Habtour, G.S. Drake, A. Dasgupta, M. Al-Bassyiouni, C. Choi, Improved reliability
testing with multiaxial electrodynamics vibration, Proceedings - Annual Reliability and
Maintainability Symposium, Article ID 5448050, 2010.
[30] C. J. Dodds and C. H. Ward, The ubiquitous four-poster, Engineering Integrity Society,
vol. 16, no. 1, pp. 1723, 2005.
[31] C. M. Awate, S. M. Panse, and C. J. Dodds, Validation of an accelerated test on a four-
post road simulator, Paper 2007-26-070, Society of Automotive Engineering, pp. 761
767, 2007.
[32] ISO 2631. Mechanical vibration and shock-evaluation of human exposure to whole-body
vibration, second ed. ISO International Organization for Standardization, 1997.
[33] R. M. French, R. Handy, and H. L. Cooper, A comparison of simultaneous and
sequential single-axis durability testing, Experimental Techniques, vol. 30, no. 5, pp.
3237, 2006.
[34] E. Habtour, C. Choi, M. Osterman, and A. Dasgupta, Novel approach to improve
electronics reliability in the next generation of US army small unmanned ground vehicles
under complex vibration conditions, Journal of Failure Analysis and Prevention, vol. 12,
no. 1, pp. 8695, 2012.
[35] D. Gabor, Theory of communication, Part 2. The analysis of hearing, Part 3. Frequency
compression and expansion, Electrical Engineers - Part III: Radio and Communication
Engineering, Journal of the Institution of Electrical Engineers, vol. 93, issue 26, p. 429-
457, 1946,.
[36] J. Ville, Thorie et Applications de la Notion de Signal Analytique, Cables et
Transmissions, vol. 2a, 1948, p. 61-74; (translate Selin I. Theory and applications of the
notion of complex signal, RAND Corporation Technical Report T-92, Santa Monica, CA,
1958).
[37] E. Wigner, On the quantum correction for thermodynamic equilibrium, Physical Review,
vol. 40, p. 749-759, 1932.
[38] K. Szabelski, The vibrations of self-excited system with parametric excitation and non-
symmetric elasticity characteristics. J Theor Appl Mech, 29, pp. 5781, 1991.
[39] S. Li, S. Yang, W. Guo, Investigation on chaotic motion in histeretic non-linear
suspension system with multi-frequency excitations. Mech Res Commun, 31, pp. 22936
2004.
[40] G. Litak, et al. Chaotic vibration of a quarter-car model excited by the road surface
profile. Communications in Nonlinear Science and Numerical Simulation, 13, pp. 1373
1383, 2008.
[41] B. Xiao, et al. Decoupling analysis on nonlinear system based on the modified
generalized frequency response functions. Mechanical Systems and Signal Processing,
42, pp. 283-299, 2014.
[42] G. Kerschen, K. Worden, A.F. Vakakis, et al. Past, present and future of nonlinear system
identification in structural dynamics. Mechanical Systems and Signal Processing, 20, pp.
505592, 2006.
[43] Work Package WP4W Report: Review of methods for evaluating human exposure to
whole body vibration. Appendix W4A to Final Report, EC Biomed II concerted action
BMH4-CT98-3291, May 2001.
[44] R.A. Lee, F. Pradko, Analytical Analysis of Human Vibration, SAE paper 680091, 1968.
[45] A. Praus, B. Powaka, K. Marchelek, et al. Active vibration control in milling flexible
workpieces, Journal of Vibration and Control, vol. 19, issue 7, pp. 1103-1120, 2013.
[46] M. H. A. Bonte, A. De Boer, and R. Liebregts, Prediction of mechanical fatigue caused
by multiple random excitations, in Proceedings of the ISMA Conference, pp. 697708,
September 2004.
[47] M. H. A. Bonte, A. de Boer, and R. Liebregts, Determining the von Mises stress power
spectral density for frequency domain fatigue analysis including out-of-phase stress
components, Journal of Sound and Vibration, vol. 302, no. 1-2, pp. 379386, 2007.
[48] D. A. Thomas, K. Avers, and M. Pecht, The trouble not identified phenomenon in
automotive electronics, Microelectronics Reliability, vol. 42, no. 4-5, pp. 641651,
2002.
[49] M. Griffin, Discomfort from feeling vehicle vibration, Vehicle System Dynamics, Vol. 45,
Nos. 7-8, p. 679-698, 2007.
[50] M. Griffin, Handbook of Human Vibration. New York: Academic Press Ltd, 1990.
[51] J. Wicher, D. Wickowski, Influence of vibrations of the child seat on the comfort of
childs ride in a car, Eksploatacja i Niezawodnosc Maintenance and Reliability, Vol.
1(77), p. 77-93, 2010.
[52] R. Burdzik, Identification of sources, propagation and structure of vibrations affecting
humans in means of transport based on the example of automotive vehicles, JVE Book
Series on Vibroengineering Vol. 1, JVE International Ltd., Kaunas, Lithuania, 2014.
[53] R.H. Rand, Lecture Notes in Nonlinear Vibrations, Published on-line by The Internet-
First University Press, 2012.
[54] A. Bernstein, RH. Rand, Delay-Coupled Mathieu Equations in Synchrotron Dynamics,
Journal of Applied Nonlinear Dynamics, Vol. 5(3), p. 337348, 2016.
[55] P. Frank Pai et al., Nonlinearity identification by time-domain-only signal processing,
International Journal of Non-Linear Mechanics, Vol. 54, p. 8598, 2013.

Highlights:
- Novel method for research on exposure to nonlinear vibration of vehicle is proposed
- time function for the evaluation of vibration duration of exposure is calculated
- Influence of damping of shock absorber on exposure to vibration is discussed